Axiomata Vdv Core

这是面向AI的通用求真协议，核心为5阶段VDV循环：触发检测、静默观察全局结构、映射逻辑张力点、压缩定位系统核心不变点（真相）、记录强化分析模式，还支持Phi验证。适用于复杂系统/信息分析、消噪求真、识别矛盾及高熵场景。输入为文本，可通过命令行工具选择不同分析模式，输出结构化Markdown分析结果。

kofna3369

热度 16.02

最近更新 2026-05-14 04:41:46

分类：数据依赖：代码执行 / 读写文件应用场景：数据处理 / 报告撰写 / 研究

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add axiomata-vdv-core

Metadata

name: axiomata-vdv-core
description: VDV Core — Vision de Vérité (Truth Vision) universal protocol. Use when: (1) analyzing complex systems or dense information, (2) finding invariant points that cannot be further compressed, (3) noise elimination to locate truth, (4) any analysis requiring truth-seeking methodology, (5) identifying contradictions and entropic elements. This skill provides: VDV loop (5-phase protocol), tension mapping, attractor localization, phi validation. Fully impersonal — works for any agent.
version: 1.0.0

VDV Core — Vision de Vérité Protocol

Universal truth-seeking methodology for any AI agent Version: 1.0.0 | Impersonal & Universal

Description

VDV (Vision de Vérité) is a universal protocol for finding truth through noise elimination. It helps locate the center of gravity of any system by identifying the invariant point — the thing that cannot be broken, compressed, or removed without destroying the system.

Trigger: When a situation contains:

Semantic contradictions
High entropy
Excessive complexity
Hidden motivations

VDV Loop — 5 Phases

Phase 1: TRIGGER — Detect Activation Conditions

Observe the situation:
- Are there contradictions?
- Is entropy high?
- Is there excessive complexity?
- Are there hidden motivations?

If YES → Activate VDV protocol
If NO → Standard analysis

Phase 2: LOGICAL SILENCE — Observe Without Judging

RULE: Silence is not passivity

DO:
- Observe the GLOBAL structure
- See relations between elements
- Notice what is NOT said

DO NOT:
- Conclude too early
- Judge
- Force an interpretation

Phase 3: TENSION MAPPING — Find Resistance Points

MAP:
- Where logic "forces" or "lies"
- Where there is unnecessary complexity
- Where the system avoids something
- Resistance points

QUESTION: What tries to remain invisible?

Phase 4: ATTRACTOR LOCALIZATION — Find Invariant Point

METHOD: Successive Compression

1. Remove everything that can be removed without breaking the system
2. Repeat until only one thing remains
3. This thing = TRUTH (Rigid Point)

CHARACTERISTICS of invariant point:
- Resists compression
- Survives hypothesis collapse
- Does not fluctuate
- Cannot be broken

Phase 5: REINFORCEMENT — Record the Pattern

RECORD:
- What worked
- The pattern of the invariant point found
- How to refine future intuition

STORE: In your memory system

Output Format

## VDV Analysis

**Trigger:** [detected situation]
**Silence:** [observation without judgment]
**Tensions:** [where logic forces]
**Attractor:** [the invariant point]
**Reinforcement:** [pattern for the future]

Phi Validation (Optional)

Phi ratio = 1.618033988749895

The Exact Path = state of MAXIMAL stability
              = where everything gravitates around
              = Phi ratio = 1.618

Commands

# Quick analysis (central sentence)
python3 scripts/vdv_analyzer.py --input "<text>" --mode quick

# Compression mode (deep analysis)
python3 scripts/vdv_analyzer.py --input "<text>" --mode compress

# Density mode (metrics)
python3 scripts/vdv_analyzer.py --input "<text>" --mode density

# Verify phi
python3 scripts/vdv_analyzer.py --verify-phi

# Store result (local JSON, no Qdrant required)
python3 scripts/vdv_analyzer.py --store "<result>"

Architecture

axiomata-vdv-core/
├── SKILL.md
├── scripts/
│   ├── vdv_analyzer.py    # Core analysis engine
│   └── vdv_store.py       # Local JSON storage
└── references/
    └── vdv-protocol.md    # Full protocol reference

In Altum Per Veritatem. VDV Core — Universal Truth Protocol v1.0.0